The Nokia difference
Network connected cameras are one of the most prolific sources of IoT data and have the potential to be among the most insightful. But companies are now faced with a deluge of video — more video than human operators can reasonably monitor. By running machine learning analytics on the camera feeds and sending to human operators just those scenes and events that contain relevant information, the full benefits of video surveillance can be realized in a wide variety of industrial and public settings.
Detect anomalous activity
Nokia uses machine learning for video analytics to detect anomalous behavior across video frames. Our models run at the edge, spotting unexpected movement, excess dwell time, and other unusual vectors. We can capture the suspect video and send it to human operators for review, or pipe it to other models for further analysis such as object detection. Scene analytics solutions dramatically reduce the amount of video requiring review, increasing the detection of real problems.
Optimize site surveillance
Nokia Scene analytics provides real time alerts for unauthorized entry into remote facilities, forensic evidence for unusual events, and because it runs on edge computing resources, can greatly reduce the bandwidth required at remote sites with limited connectivity.
Nokia Scene analytics can detect and alert supervisors when personnel or equipment enter unsafe locations in industrial settings, or heavy machinery is out of position creating a hazard. It can detect people, cars, and objects stuck at railroad crossings or potential criminal activity at rail stations and yards.
Scene analytics optimizes video surveillance for: